Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/iamlukethedev/claw3d/agents-mdgit clone --depth 1 https://github.com/iamlukethedev/Claw3DWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/iamlukethedev/claw3d/agents-md)<a href="https://agentmods.dev/instructions/iamlukethedev/claw3d/agents-md"><img src="https://agentmods.dev/badge/instructions/iamlukethedev/claw3d/agents-md.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00608 | $0.00608 |
| Opus 5 | $0.00304 | $0.00304 |
| Sonnet 5 | $0.00122 | $0.00122 |
| Haiku 4.5 | $0.00061 | $0.00061 |
Grade A, and why
Claw3D AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 42 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Instructions
Keep repository instructions generic and safe for open source.
This repo is a frontend for OpenClaw. Keep any OpenClaw runtime checkout separate from this repository.
Do not modify the OpenClaw source code. When the user asks for changes, they are asking for changes to this app. Your solutions should be applied to this app but to understand the full context of implementing your solution, you will need to search through OpenClaw's source code.
If you use local private overlay instructions, keep them outside the repository and do not commit them here.
Do not commit personal, environment-specific, or secret instructions to this repository.
Cursor Cloud specific instructions
Service overview
Claw3D is a Next.js 16 frontend (TypeScript, React 19, Three.js, Phaser) for OpenClaw. It runs a custom Node.js server (server/index.js) that bundles a same-origin WebSocket proxy to the upstream OpenClaw Gateway. No database or Docker is required. The only hard system dependency is Node.js 20+ with npm 10+.
Running the app
npm run devstarts the dev server on port 3000 via the custom server (node server/index.js --dev).- The app requires a running OpenClaw Gateway to show agent data. Without one, the UI loads but shows the gateway connection form. This is expected and not an error.
.envis copied from.env.example; seeREADME.md"Configuration" for variable descriptions.
Lint, typecheck, and tests
npm run lint— ESLint. The codebase has a small number of pre-existing warnings and one pre-existing error (inRetroOffice3D.tsx).npm run typecheck—tsc --noEmit. Pre-existing type errors exist in some test files (agentChatPanel-*.test.ts) due to a staleonOpenSettingsprop.npm run test -- --run— Vitest unit tests (use--runfor single-run mode). A few pre-existing failures exist.npm run e2e— Playwright E2E tests; requiresnpx playwright installfirst.npm run smoke:dev-server— starts the dev server on a random port and verifies HTTP response.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 42 lines · 608 tokens per session scan A 4510e6e48687
Claw3D AGENTS.md is an instructions file published in the GitHub repository iamlukethedev/Claw3D (2,200 stars, last pushed 13d ago), licensed MIT. It adds 608 tokens to every session, about $0.0030 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
Hermes3D AGENTS.md
Instructions for iamlukethedev/Hermes3D, covering agent instructions, development environment instructions, service overview, running the app and lint, typecheck, and tests.
terminal AGENTS.md
Instructions for m4tt72/terminal, covering agents.md - repository guide for agentic coding, project overview, commands, development and testing.
p5.js AGENTS.md
AGENTS.md instructions for processing/p5.js, covering detailed guidelines for ai coding assistants and agents, generating any analysis of the codebase, stop signs for agents, character and tone and who you are in this conversation.
skyrim-community-shaders copilot-instructions.md
Copilot instructions for community-shaders/skyrim-community-shaders, covering github copilot instructions, primary documentation, project overview, environment and build essentials and windows-only requirements.
pdflokal CLAUDE.md
Claude Code instructions for ojanlubis/pdflokal, covering pdflokal — app/ (the bench: all code execution), hard constraints, the tree, the seo generator's split is load-bearing and tokens and theming.
reorder AGENTS.md
AGENTS.md instructions for reorder-js/reorder, a project described as: Open Source Medusa.js Subscription Plugin.